Evaluation of empirical models for predicting methane emissions of dairy cattle from Latin American studies.

CABEZAS E. ; GARCÍA R. ; HUANCA N.; GUALDRÓN-DUARTE, LAURA BIBIANA; ORELLANAS G. ; RIVERA F.; SANCA Y.; VELÉZ V.; GERE J. I.

Florida

Conferencia; 8th Greenhouse Gas and Animal Agriculture Conference.; 2022

Dairy cattle in Latin America may have constraints to express their genetic potential and this may reflect in lower feed efficiency and likely increased methane (CH4) emissions when compared with production records of animals raised in other latitudes. The objective of this study was to evaluate the performance of ten empirical models of varying complexity (including both animal and diet-related factors) for predicting total CH4 emissions (g/day), with this being measured using the SF6-tracer technique. Data were collected (means by treatment) from peer-reviewed literature studies (n=13), mostly conducted under grazing conditions (n= 10). Both lactating and non-lactating females (n=202) were included in the dataset with 69.3% of these being Bos Taurus (mainly Holstein), 24.9% crossbred B. taurus x B. indicus, and 5.9% B. indicus (Gyr). Average body weight (BW) was 474±100 kg, dry matter intake (DMI)=14.5±4.49 kg/d, total CH4 emissions=337±146 g/day, and milk yield=20.2±4.68 kg (10 studies). Mean concentrate proportion offered to the animals was on average 12% of the diet on an as fed basis. Model evaluation was conducted using mixed model regression. Observed values were adjusted for the random effect of study. Mean biases were evaluated by the deviation of regression intercepts from zero while the deviation of the slopes of the regression equations from unity was used to determine the presence of linear biases. Model accuracy was assessed by the calculation of the root mean square prediction error (RMSPE). Among evaluated models, the non-linear equation proposed by Mills et al. (2013), which includes DMI as single prediction factor, displayed the best performance in the model ranking (lower RMSPE = 35.8 g of CH4/day) and slightly under predicted total CH4 emissions by 3.39 g/day. The results of this study confirmed DMI as a key factor driving enteric CH4 emissions in dairy cattle.